You are on page 1of 11

Archives of Biochemistry and Biophysics 602 (2016) 95e105

Contents lists available at ScienceDirect

Archives of Biochemistry and Biophysics

journal homepage:

Transfer RNA: From pioneering crystallographic studies to

contemporary tRNA biology
Pablo Ferna n a, Ce
dric Schelcher b, Joseph Chihade a, c, Benot Masquida d,
 b, Claude Sauter a, *
Philippe Giege
UPR 9002 du CNRS, Architecture et Reactivit
e de lARN, IBMC, Universite de Strasbourg, 15 rue Ren
e Descarte, Strasbourg, F-67084, France
UPR 2357 du CNRS, IBMP, Universite de Strasbourg, 12 rue du general Zimmer, F-67084, Strasbourg, France
Department of Chemistry, Carleton College, 1 North College Street, Northeld, MN 55057, USA
UMR 7156 du CNRS, Genetique Moleculaire, Genomique, Microbiologie, Universite de Strasbourg, 4 All
ee Konrad Roentgen, Strasbourg, F-67084, France

a r t i c l e i n f o a b s t r a c t

Article history: Transfer RNAs (tRNAs) play a key role in protein synthesis as adaptor molecules between messenger RNA
Received 3 November 2015 and protein sequences on the ribosome. Their discovery in the early sixties provoked a worldwide
Received in revised form infatuation with the study of their architecture and their function in the decoding of genetic information.
29 February 2016
tRNAs are also emblematic molecules in crystallography: the determination of the rst tRNA crystal
Accepted 3 March 2016
structures represented a milestone in structural biology and tRNAs were for a long period the sole source
Available online 8 March 2016
of information on RNA folding, architecture, and post-transcriptional modications. Crystallographic data
on tRNAs in complex with aminoacyl-tRNA synthetases (aaRSs) also provided the rst insight into
Transfer RNA
protein:RNA interactions. Beyond the translation process and the history of structural investigations on
Crystallography tRNA, this review also illustrates the renewal of tRNA biology with the discovery of a growing number of
Protein synthesis tRNA partners in the cell, the involvement of tRNAs in a variety of regulatory and metabolic pathways,
Genetic code and emerging applications in biotechnology and synthetic biology.
Translation 2016 Elsevier Inc. All rights reserved.
RNA:protein recognition

1. Introduction were isolated along with an essential family of enzymes,

aminoacyl-tRNA synthetases (aaRSs), which catalyze the specic
Since the mid-1950s crystallography has accompanied the birth coupling of amino acids to their tRNA carriers [5].
and the development of molecular biology, providing biologists At this point, it became crucial to gain an insight into the
with pictures of their favorite biomolecules and helping them structural properties of these molecules in order to understand the
describe biological processes at an atomic scale [1]. The determi- specicity of substrate recognition and of catalytic mechanisms
nation of the iconic three-dimensional structure of the DNA double ensuring the delity of tRNA aminoacylation reactions and, by
helix by Watson and Crick revolutionized the understanding of life extension, of protein synthesis. This triggered a worldwide effort to
[2]. It led Crick to formulate his central dogma explaining the decipher the crystal structure of tRNAs and related enzymes. In
propagation of genetic information by DNA replication, transcrip- 1966 the sequencing of tRNAAla from yeast revealed complemen-
tion into RNA and translation into proteins, following a genetic code tary stretches of nucleotides suggesting a 2D folding resembling a
that associates every base triplet or codon in the DNA/RNA cloverleaf [6]. The next three sequences of tRNATyr, tRNAPhe, and
sequence with an amino acid in the polypeptide chain [3,4]. Yet an tRNASer [7e9] reinforced the idea of a four-arm organization. They
essential piece of the puzzle was missing and Crick proposed the also led to the concept of anticodon base triplet complementary to
'adaptor hypothesis' in 1955, postulating the existence of a dedi- the codon of the carried amino acid, establishing a physical link
cated class of molecules making the link between nucleic acid and between nucleic acid and protein alphabets/sequences. Hence, it
protein sequences [3]. Soon after a family of small RNAs, rst called became clear that all tRNAs should share a common scaffold to
soluble RNAs (sRNAs) and quickly renamed transfer RNAs (tRNAs), ensure their recognition by the ribosome and by elongation factors,
while also expressing a specic identity with regard to their
cognate aaRSs. The quest for structural data led to a rst important
* Corresponding author. breakthrough in the mid-seventies with the determination of the
E-mail address: (C. Sauter).
0003-9861/ 2016 Elsevier Inc. All rights reserved.
96 ndez-Milla
P. Ferna n et al. / Archives of Biochemistry and Biophysics 602 (2016) 95e105

crystal structure of yeast tRNAPhe by two groups in the US and one boomerang-shape proposed based on early small angle X-ray
in the UK [10e12], followed by that of yeast tRNAAsp in France [13]. scattering (SAXS) measurements [21]. This canonical 2D cloverleaf
These crystallographic studies revealed the characteristic L-shape and 3D L structure were later conrmed in a dozen crystal struc-
of adaptor molecules. The story continued with the determination tures of free tRNAs, as well as by tens of thousands of tRNA se-
of the rst crystal structures of tRNA:aaRS complexes in the early quences derived from a growing number of genomes and collected
nineties reviewed in Ref. [14] and culminated in the 2000s with the in dedicated databases such as the tRNAdb (http://trna.bioinf.uni-
description of the adaptors bound to the A, P and E sites of the [22]. Exceptions to this general pattern have been
ribosome [15]. found in animal mitochondria, where tRNAs can display size re-
In the context of the celebration of a century of X-ray crystal- ductions due to shorter loops in the D or T arms, and in some cases,
lography, the aim of this review is to illustrate how early studies of the complete absence of one or two of arms (Fig. 2B). The most
the key protagonists in the translational machinery such as tRNAs extreme case is a functional tRNA consisting of only 42 nucleotides
have contributed to many aspects to structural biology and how (to be compared with usual sizes of 70e95 nt) recently described in
vital this tRNA research continues to be. Solving the initial struc- the mitochondria of the worm Romanomermis culicivorax [23]. This
tures represented a considerable challenge and required method- raises questions about the folding and the stability of such tRNAs
ological developments in sample preparation, crystallization and and their recognition by cellular factors including the ribosome.
crystallography. For almost 15 years, tRNAs were the sole source of Another structural characteristic of tRNAs is the presence of
structural information about RNA and provided the basis for the numerous post-transcriptionally modied nucleosides (Fig. 2C).
understanding of RNA architecture and function, as well as infor- Besides methylated bases, the rst sequence of yeast tRNAAla also
mation about RNA:protein recognition and the role of RNA modi- revealed the presence of dihydrouridines in one loop, as well as a
cations. Finally, while tRNAs entered textbooks in the nineties as ribo-thymidine (T) and a pseudouridine (J) in another, leading to
central actors in translation [16], a number of functional questions the current names of D- and T- (or TJC-) loops/stems as these
have recently emerged from the genome-wide examination of modications are almost universally conserved. Inosine was also
tRNA diversity. These include the discovery of new tRNA partners found at position 34, soon identied as the rst position of the
and related pathways and the observation of tRNA mutations anticodon, which base pairs with the third position of the codon in
related to human pathologies, stimulating novel research directions the messenger RNA. Following the 'wobble hypothesis' of Crick [3],
on tRNA function, structure, dynamics, and biotechnology. an inosine at this position in the tRNA extends its decoding capacity
to the synonymous codons ending by A, C or U (GCA, GCC and GCU
2. From rst crystal structures to the -omics era in the case of tRNAAla with an IGC anticodon). For this reason, the
number of tRNA species in a cell is always lower than the 61 codons,
A rst step towards the crystallographic analysis of tRNAs was and can be as low as 22 in mammalian mitochondria, due to the
the separation of different species from natural sources like baker's presence of specic modications at positions 34 and 37 of the
yeast. RNA is distinguished from DNA by the presence of a hydroxyl anticodon loop [24]. Modied nucleosides can serve as recognition
group in 20 position of the sugar which confers both exibility and signals for cognate synthetases (see below) and are known to in-
reactivity, making RNA molecules much more labile and difcult to crease the stability of the tRNA scaffold, as reviewed in Ref. [25]. A
work with than DNA. In the sixties the method of countercurrent classic example of modication that affects tRNA structure is the
distribution, based on partitioning tRNAs according to their hy- m1A at position 9 in human mitochondrial tRNALys which drives
drophobicity in aqueous and organic solvents, played a key role. the formation of the functional cloverleaf over an alternative
This technique allowed the preparation of pure samples on the hairpin structure [26]. More than a hundred such RNA modica-
hundreds of milligram scale required for biochemical character- tions have been identied (Fig. 2C) and their structures and posi-
ization, sequencing and eventually crystallization [17]. The intro- tions in tRNAs, together with information on their biosynthetic
duction of the vapor diffusion method was also crucial for the pathways, have been cataloged in the RNA modication (http://rna-
success of tRNA crystallization [18]. Other micromethods, such as and the Modomics (http://modomics.
dialysis or free-interface diffusion, were developed to crystallize databases [27,28].
aaRSs and aaRS:tRNA complexes because they were better adapted
to sensitive biological samples and have since been widely adopted. 3. Exploring the mechanisms of tRNA:synthetase recognition
A new discipline called biocrystallogenesis, aiming at better un-
derstanding and mastering the process of biological crystallization, The rst structural information on tRNA:synthetase interactions
directly emerged from the difculties encountered in the prepa- came more than a decade after the solution of tRNAPhe structure.
ration of crystals suitable for diffraction studies [19]. These efforts The delay was mainly due to the difculties in the preparation of
eventually yielded tRNA crystals diffracting at resolution of ~3 at homogeneous complexes for crystallization and crystal quality
the turn of the seventies. optimization. The rst structures of tRNA:aaRS complexes specic
Today, when it has become possible to determine X-ray struc- for glutamine and aspartate [29,30] highlighted that, although
tures on a simple laptop computer, it is difcult to imagine how synthetases catalyzed the same type of reaction, they diverge in
solving a tRNA structure once represented long and tedious work, terms of sequence and structure. The aaRS family was divided into
collecting diffraction pattern on photographic lms and building a two classes based on the topology of their catalytic domain [31]: a
model manually in hand-drawn electron density maps (Fig. 1). Rossmann fold, with its characteristic parallel b-sheet, is found in
These pioneering crystallographic studies led to the well-known L- monomeric Class I aaRSs, while Class II enzymes are generally
shape model. The building of full atomic models revealed how dimeric and are built around an antiparallel b-sheet. The determi-
helical domains of the cloverleaf get stacked pairwise (acceptor and nation of a number of crystal structures of tRNA:aaRS complexes
T helices, D and anticodon helices; Fig. 2A) to form the two arms of has revealed the specicities and dynamics of class I and class II
the L, held together by a network of tertiary interactions predicted aaRS binding to tRNA.
from sequence conservations by Levitt in 1969 [20]. Although the The observation of close contacts between the partners,
angle between the arms in tRNAAsp is more open than in tRNAPhe together with mutagenesis experiments on both partners, rein-
(100 instead of 90 ), a sign of the intrinsic exibility of tRNA forced the concept of a second genetic code. This operational code
backbone, both tRNAs adopt a similar fold, in agreement with the involves tRNA determinants and tRNA anti-determinants, which
P. Ferna n et al. / Archives of Biochemistry and Biophysics 602 (2016) 95e105 97

Fig. 1. Determination of yeast tRNAAsp crystal structure [13]. Structural investigations on yeast tRNAAsp started in the seventies and illustrate well how crystallographic studies
were performed at the time. (A) Diffraction patterns were collected on photographic lms from which reection intensities were quantied. Multiple isomorphous replacement
with heavy atoms was used to determine sets of phases and to compute slices of electron density maps (B). A low resolution model in balsa wood was rst assembled (C), then the
tRNA backbone was built in metal wire by connecting the positions of phosphates, which were easily identied in the density (D). Using an optical device called Richards box (see and electron density maps manually transferred on glass windows, the French research team at IBMC, Strasbourg (E) could
, M.-B. Comarmond, D.
build a wire frame atomic model (scale: 2 cm 1 ) showing the characteristic L-shape of yeast tRNAAsp. From the left to the right: J.-C. Thierry, R. Giege
Moras (courtesy of R. Giege

are chemical signals dening the identity of a family of tRNAs with isolated aaRS structures but no structures of complexes [37].
respect to their cognate aaRS or preventing the interaction and Recently, mitochondrial systems have drawn substantial interest
misacylation by a non cognate enzyme, respectively [32]. As might due to the discovery of mutations in tRNAs and aaRSs associated
be expected, the anticodon region often constitutes a major identity with a variety of human pathologies [39,40]. Future investigations
element. In several complexes, anticodon nucleotides are distorted will be necessary to examine how these mutations affect expres-
so as to be bound in specic enzymatic pockets (see, for example, sion, folding, binding, catalysis or other functions of tRNAs aside
the structure of the GlnRS-tRNAGln complex [33]). from aminoacylation and translation.
Identity elements are also frequently found in the acceptor stem,
but the interpretation of specic protein contacts with this region 4. tRNA as template for RNA modeling
of the tRNA is more difcult, since interactions with both the
ribose-phosphate backbone and nucleotide bases may play a role. As mentioned above, the rst tRNA structures were, for a long
The clearest example may be provided by two recent structures of period, the only examples of complex RNA architecture (Fig. 3). As
A. fulgidus alanyl-tRNA synthetase [34]. In one, the enzyme is such, they were instrumental in elaborating how double-stranded
complexed with a cognate tRNA, containing the G3:U70 wobble helices can be connected by structured loops in which all nucleo-
pair, which is an identity element, while the second is a complex tide edges (Watson-Crick, Hoogsteen, sugar) are used in base-
with an A3:U70 tRNA mutant. The structures reveal a multifaceted pairing. Important roles of the 20 hydroxyl group of nucleotides
set of protein-RNA interactions which direct the A76 nucleotide of were noted, such as its capacities to constrain the ribose pucker,
the cognate tRNA into the aminoacylation active site, but distort the forcing A-form RNA helices, and to simultaneously act as a donor
backbone of the mutant tRNA and prevent A76 from attaining the and acceptor of hydrogen bond. Despite their relatively limited size
proper conguration for aminoacylation. Other tRNA:aaRS struc- and number, as the only available templates for RNA modeling, the
tures have demonstrated idiosyncratic modes of tRNA recognition tRNA crystal structures have long been a source of inspiration.
that are difcult to describe within the framework of identity ele- The observation of an overall L-shape architecture stabilized by
ments, such as T. thermophilus SerRS recognition of the long vari- pairwise helix stacking led to the concept of RNA domain assembly.
able arm of tRNASer [35] and shape-selective contacts between This assembly is further stabilized by loop-loop interactions on one
E.coli CysRS and tRNACys [36]. side and base triples on the other (Fig. 4). The stunning beauty of
These structural studies also shed light on the high degree of the structures was due to the harmony of the backbone curves
tRNA plasticity in tRNA:aaRS interactions [37]. For example, in the resulting from interactions between bases in a layout so natural
case of yeast aspartylation system, the comparison of tRNAAsp that the whole appeared to be unconstrained. As masterpieces
structure with the structure of its complex with AspRS revealed resulting from millions of years of molecular adaptation, the tRNA
important conformational changes altering its structure upon crystal structures inspired structural biologists. The observation
enzyme binding, i.e. with the two arms of the overall L shape of that the single strands joining the helical segments adopted precise
tRNAAsp presenting a reduced angle [30]. In class I aaRSs including architectures led to the idea that they could constitute individual
an editing domain, the 30 end of the tRNA can swing from the structural motifs.
aminaocylation site to the editing site to allow hydrolysis of an This essential RNA structural repertoire (See Fig. 4) could in
incorrect amino acyl ester [38]. Despite the rapidly growing num- principle be used as a building set to generate theoretical RNA
ber of crystal structures of tRNA:aaRS complexes, the existing data structural models by pasting together individual building blocks.
remain highly biased in favor of bacterial systems, with few However, in the eighties the organization of the secondary struc-
archaeal and eukaryotic structures available. Knowledge of organ- tures of ribosomal RNAs (rRNA) already indicated that the RNA
elle aminoacylation systems is even more scarce, with only a few structural diversity was not exhaustively contained in the tRNA
98 ndez-Milla
P. Ferna n et al. / Archives of Biochemistry and Biophysics 602 (2016) 95e105

Fig. 2. tRNA structural organization. (A) 2D and 3D structures of tRNAPhe (PDBid 1TN1). Tertiary interactions that stabilize the L-shape are depicted on the cloverleaf as broken
lines (more details in Fig. 4). The arms of the cloverleaf are depicted with the same colors throughout the gure. (B) Examples of variations on the classical 2D organization with
typical armless mitochondrial tRNAs and the tRNA-like structure (TLS) found in 30 of the genomic RNA of Turnip Yellow Mosaic Virus (TYMV). (C) Distribution of post-transcriptional
modications on a tRNA cloverleaf with conventional nucleotide numbering. (D) General nomenclature of tRNA-derived fragments (tRFs). Mature tRNAs and precursors can be
cleaved at different positions to produce various tRFs. The atomic model of tRNAPhe was rendered using PyMOL (Schro dinger), as well as atomic models in Fig. 4 and 5.

structures. The number of loop residues well illustrates this idea. repertoire by ~7 fold and revealed a wealth of magnicent and
The tRNAs display seven- or eight-membered loops, while rRNAs daring RNA folds, RNA-RNA and RNA-protein interaction motifs.
contain less or more nucleotides in the loops closing hairpins. Strikingly, most of these folds are recurrent, like the sarcin or loop E
Moreover, rRNAs present internal loops or complex junctions motif, the GNRA and UNCG tetraloops, kink-turns [48,49], C-loops
linking two to ve helical segments impossible to map onto tRNA [50], UA handles [51], and A-minor motif [52,53]. For example, the
subdomains. At this stage, it was difcult to anticipate how the RNA A-minor motif is very frequent due to the versatility of the O20
structural repertoire would expand. However structural biologists group interactions, although it cannot be deduced from compara-
could understand that the diversity of base-pairs beyond the tive sequence analysis [54]. Some of these motifs are even remi-
Watson-Crick classics would be at play. Additional likely in- niscent of tRNA: one can nd a fair representation of anticodon
teractions were observed by analysis of tRNA crystal packing. Thus, loops, U-turns, as in the structure of the lariat-capping ribozyme
building unknown motifs from interactions between nucleotides [55], and T-loops, the latter being crucial for tRNA recognition by
became a eld of research. All atom modeling independently added RNase P [56], T box regulators [57] and ribosome [15,58].
stereochemical restraints on top of the secondary structure con- RNA modeling has never been as necessary as today since
straints, conferring signicant predictive power to the method. massive numbers of RNA sequences are being identied by new
Structural biologists could use RNA helices, base triples and kissing generation sequencing (NGS) methods at a rate far faster than
loops from the tRNA structures to build complex architectures like conventional structural methods such as crystallography can follow
group I introns [41]. up. The continued efforts of structural biologists to increase the
The number of available RNA crystal structures increased number of RNA crystal structures are thus mandatory to approach
gradually (Fig. 3), each time revealing new structural features that completion of the RNA structural repertoire which makes up the
helped structural biologists to better understand RNA architecture modeling building set. The identication of both secondary struc-
and folding principles [42e44]. However the mist really cleared tures and nucleotides in proximity in RNAs by high-throughput
upon the appearance of the crystal structures of the ribosomal chemical probing methods can then be coupled to automatic
subunits [45e47], which multiplied the size of the structural modeling methods based on the adjunction of RNA fragments. This
P. Ferna n et al. / Archives of Biochemistry and Biophysics 602 (2016) 95e105 99

Fig. 3. tRNA and RNA entries in the PDB (as of December 2014). The plot bar shows the number of PDB entries per year for structures of tRNA alone (red), tRNA:protein complexes
(blue, this group includes tRNA:ribosome complexes), all RNA:protein complexes (green) and all isolated RNAs including tRNAs (violet). The inset gives a close-up view for the years
1978e1993. Until 1994 and the determination of rst ribozyme structures, tRNAs were the exclusive source of information on the folding and architecture of complex RNA
molecules. Then the study of the ribosome led to a burst in RNA and tRNA/RNA:protein complex data in the PDB.

type of approach, exemplied by MOHCA-seq coupled to modeling several additional modication and editing events [71]. A number
with Rosetta [59] have proven to reach subhelical resolution (10 ). of structures of tRNA modifying enzymes in complex with complete
tRNA have exposed several stages of the tRNA modication process,
5. tRNA partners in the cell including initial RNA binding, the formation of covalent in-
termediates and the release of the modied RNA reviewed in Ref.
As a central actor of the translation machinery, many of tRNA's [72]. The structures of enzyme-tRNA complexes allowed tRNA
partners are involved in the process of protein synthesis, including modifying enzymes to be categorized in distinct classes depending
aaRSs, initiation and elongation factors that deliver aminoacyl- on their requirement for the L-shape structure of tRNA for RNA
tRNAs to the ribosome. A growing number of functional com- recognition. Group I enzymes bind either nucleotides in the anti-
plexes not directly related to protein synthesis have also been codon loop or residues embedded inside the tRNA structure,
structurally characterized over the years, including maturation whereas group II enzymes require the three dimensional structure
enzymes involved in tRNA biogenesis, regulatory T-boxes in mRNA of tRNA for recognition [73]. The last step in tRNA maturation is the
and tRNA-like structures in viral RNAs as shown in Fig. 5, illus- addition of a CCA 30 group by CCA adding enzymes (CCases).
trating the diversity of cellular functions associated with tRNAs. Structural analyses have revealed that CCases use a template in-
dependent RNA polymerization mechanism [68]. More recently,
5.1. tRNA biogenesis the determination of structures of archaeal CCA adding enzymes in
complex with tRNA has revealed how the enzymes discriminate
Like many other RNA molecules, tRNAs are expressed as pre- between the incorporation of C or A at position 76. The discrimi-
cursor transcripts and need to undergo several maturation steps in nation against incorporation of C at this position arises because
order to become functional. tRNA precursors are rst cleaved by protein exibility results in the improper geometry of CTP in the
two endonucleases called RNase P and RNase Z, which remove 50 active site [74].
leader and 30 trailer sequences of tRNA precursors respectively. The
RNase P function was believed to be ubiquitously performed by 5.2. tRNAs in the ribosome
ribonucleoproteins containing a ribozyme (P RNA) until the recent
discovery of protein-only RNase P enzymes in eukaryotes reviewed Aminoacylated tRNA in complex with initiation and elongation
in Ref. [69]. The determination of the structure of the bacterial factors reaches the ribosome where the incorporation of amino
Thermotoga maritima RNase P in complex with tRNA has revealed acids into the emerging protein chain through peptide bond for-
that tRNA-P RNA recognition occurs through shape complemen- mation is achieved. A rapidly increasing number of high-resolution
tarity, specic intermolecular contacts and base-pairing in- structures of tRNA in complex with ribosome, ribosomal subunits
teractions [56]. Interestingly, although a structure of protein-only and/or elongation factors at different stages of translation have
RNase P in complex with tRNA has not yet been solved, initial been obtained by both X-ray crystallography and cryo-electron
biochemical and biophysical data suggest that protein-only and microscopy (cryoEM) [75]. These structures together with
ribonucleoprotein RNase P share similar processes for specic tRNA biochemical and computational approaches have revealed the
recognition [70]. After the initial end-trimming, tRNAs undergo plasticity of tRNA molecules during their interaction with the
100 ndez-Milla
P. Ferna n et al. / Archives of Biochemistry and Biophysics 602 (2016) 95e105

Fig. 4. The early observation of tRNA structures shed light on RNA folding and architecture. (A) Overall ribbon view of the tRNAPhe crystal structure (PDB ID 1TN1). The different
domains are indicated and colored individually. The variable loop lies behind the dihydrouridine stem loop (dim orange). (B) The core of the tRNA is stabilized by base triple
interactions. In the foreground, the last two base pairs of the D stem are represented, U12-A23, and C13-G22. In the deep groove of the stem the Hoogsteen edge of U9 and the Watson-
Crick edge of m7G46 interact with the Hoogsteen edge of A23 and of G22, respectively. The rst nucleotide of the D loop, A14, is stabilized by U8, which directly connects the Acceptor
stem to the D domain. The resulting trans Watson-Crick/Hoogsteen interaction is further stabilized by A21, the terminal residue of the D loop. m7G46 stacks in between A21 and G22.
As seen in (C), stacking interactions extend farther than the base triple ensemble, which is wrapped by C48 and G45. The latter interacts with the rst base pair of the D stem (not
shown), while the former stacks onto U59, the unpaired penultimate nucleotide in the 30 side of the T loop. (D) The U-turn motif consists in the interaction between the N3 imino
proton of U33 with the proximal phosphate oxygen atom of A36. This motif is observed recurrently in RNA structures, for example in the T loop of the tRNA (P55 and M1A58 in panel
F). (E) A view of the overall anticodon loop shows the arrangement of the three nucleotides 34e36 (italic) involved in mRNA recognition. Hyper modication of nucleotide 37
prevents its interaction with a fourth nucleotide on the mRNA, which would result in mistranslation. (F) A view of the nucleotides involved in the interaction between the T and D
loops depicts how stacking interactions promote close contact between nucleotides which result in exquisite hydrogen bonds. Notably, packing contacts in tRNA crystals between
anticodon loops of adjacent tRNA molecules dictate the codon-anticodon interaction observed on the ribosome. Hydrogen bonds are indicated by solid dashed lines. '*' indicates
modied nucleotides. *G46: 7-methylguanosine; *G34: 20 -O-methylguanosine; *C32: 20 -O-methylcytidine; *U54: 5-methyluridine; P55: pseudouridine; *A58: 1-methyladenosine.

ribosome. It thus becomes increasingly evident that the intrinsic to the P site of the ribosome. In all kingdoms, its anticodon stem
structural exibility of tRNA is pivotal for protein synthesis [76]. In contains 3 consecutive G-C pairs followed by an additional C32-A38
the ribosome, tRNAs bind the primary aminoacyl (A) site, travel to wooble pair [64,79]. This unique sequence leads to a specic anti-
the peptidyl (P) site and leave the ribosome at the level of the exit codon loop conformation which is proposed to facilitate the ac-
(E) site. In this process, tRNA conformational changes are required commodation of Met-tRNAMet i in the P site.
for the concerted action of the ribosome's small and large subunits In prokaryotes, the small subunit of ribosomes binds the initi-
during translation initiation, for the decoding steps where amino- ator tRNA, initiation factors and mRNA to form the 30S pre-
acylated tRNA are incorporated and for ribosome translocation [77]. initiation complex in a sequential process. After the positioning of
Translation is initiated with the binding of the initiator tRNAMet
i the mRNA, the anticodon of the initiator tRNA is base-paired with
to the P site of the small ribosomal subunit, where its anticodon the initiation codon and the sequential release of initiation factors
basepairs with the mRNA start codon. Initiator tRNAs possess induces association of the large subunit to establish the 70S initi-
structural features that distinguish them from elongator tRNAs, ation complex [80]. In contrast, in eukaryotes, the initiation of
including the elongator tRNAMet, and enable their specic recog- translation is a more complex and tightly regulated process
nition and shuttling by initiation factors IF2/aIF2/eIF2. In Escher- engaging over ten initiation factors, in which eukaryotic ribosomes
ichia coli, the initiator tRNA bears a C1-A72 mismatch (instead of a use a scanning mechanism resulting in the formation of the initi-
G1-C72 pair in the elongator) which is essential for the formylation ation complex with the initiator tRNA in its ultimate location
of the methionine. This formyl group ensuring a specic binding to [81,82]. In both eukaryotes and prokaryotes, the variety of confor-
IF2, instead of EF1A (or EF-Tu), the elongation factor [78]. In Eu- mations observed for initiator tRNAs reveals the plasticity of the
karyotes and Archaea, a weak base A1-U72 is a major identity acceptor arm and suggests that both the assembly of ribosome
element for binding to e/aIF2 whereas J54, a major recognition subunits and the integration of initiator tRNA into the P site might
signal for eEF1A, is absent. Met-tRNAMeti is shuttled by IF2 or e/aIF2 be regulated by the distinct conformations adopted by tRNA during
P. Ferna n et al. / Archives of Biochemistry and Biophysics 602 (2016) 95e105 101

Fig. 5. Structural gallery of tRNAs and their cellular partners. This selection of representative structures shows the diversity of tRNA partners and includes free tRNAs and tRNA-
like molecules with (1) the yeast phenylalanyl-tRNA [10,11], (2) the murine selenocysteinyl-tRNA characterized by a large variable region and an acceptor stem with 8 base-pairs
[60], (3) the tRNA-like structure present at the 30 of TYMV genomic RNA [61]; enzymatic complexes active in tRNA aminoacylation and protein synthesis, involving (4) a monomeric
class I glutaminyl-tRNA synthetase [29], (5) a dimeric class II aspartyl-tRNA synthetase [30], (6) the transamidosome [62], (7) the methionyl-tRNAfMet-formyltransferase [78];
bacterial and archeal elongation factors (8) EF-Tu [63] and (9) aIF2 [64], and (10) the bacterial ribosome bound to EF-Tu:tRNA [65]; tRNA maturation enzymes with (11) the
ribonucleoproteic RNase P [56] (12) the ribonuclease Z [66], (13) tRNA-guanine transglycosylase associated with tRNA in so-called lambda open conformation [67] and (14) CCA
adding-enzyme [68]; regulatory elements represented by (15) the T-box riboswitch [57]. All tRNAs are depicted with an orange backbone and pink bases, free tRNAs are repre-
sented, tRNA partners are shown with a transparent surface and different colors are used to distinguish their subunits.

the initial steps of translation [83]. associated with the rotation of the 30S relative to the 50S subunit
When the ribosome is assembled, decoding of the mRNA can along with the stalk rearrangement as shown by cryoEM studies of
start and the delivery of elongator tRNAs is ensured by the elon- prokaryote and eukaryote complexes, which revealed several tRNA
gation factors. The determination of the structure of ternary com- intermediate congurations [76,90,91].
plex made of EF-Tu, Phe-tRNAPhe and an analog of GTP represented Interestingly, in some cases, tRNAs interact with ribosomes in
an important breakthrough in the mid 90s and shed light on the ways that go beyond the primary association of tRNA as part of the
recognition of aminoacylated tRNAs [63]. Recent progress in ribo- mRNA decoding process. A tRNA was also found associated to the
some crystallography enabled the observation of this ternary large subunit of the mitochondrial ribosome in animals. This tRNA,
complex docked onto the ribosome (Fig. 5) and revealed the most likely mitochondrial tRNAPhe, is referred to as CP tRNA. It is an
structural plasticity of tRNAs [65]. Then, during the decoding pro- integral component of the central protuberance (CP) of the 39S
cess, the ribosome has to discriminate between cognate amino- subunit, where it structurally replaces the 5S ribosomal RNA, which
acylated tRNA-elongation factor complexes and non-cognate is not encoded in animal mitochondrial genomes, despite being
complexes. This is achieved through a kinetic discrimination ubiquitous in cytoplasmic ribosomes [92,93].
mechanism that allows incorrect tRNA species to be rejected at
different stages. Decoding relies on Watson-Crick pairing between
5.3. tRNA beyond translation
mRNA codons and tRNA anticodons together with 30S conforma-
tional changes [84]. The ribosome plays an active role in the ac-
In addition to their canonical function, tRNA molecules partici-
curacy of this process as it uses an ensemble of local and global
pate in a variety of other processes [37], such as cell wall synthesis
conformational changes to control the delity of tRNA selection e.g.
in bacteria (tRNASer) or the porphyrin synthesis pathway (tRNAGlu).
Refs. [85,86]. Here again the structural exibility of tRNA plays a
Viruses frequently take over some feature of tRNA biology. In
fundamental role to allow the decoding process in both prokaryotic
particular, retroviruses recruit particular host tRNAs for the priming
and eukaryotic systems [87].
of reverse transcription of their genomes [94]. Likewise, large DNA
For the translocation of ribosomes, peptide transfer takes place
viruses such as bacteriophages, phycodnaviruses, and mimiviruses
after the entry of tRNA in the A site guiding the formation of the
express their own tRNAs, most probably to regulate translation
pre-translocation complex poised to translocate tRNA-mRNA du-
during viral infection [95]. In bacteria, the T-box riboswitch differs
plexes by one codon e.g. Refs. [87,88]. In this motion, the acceptor
from all other identied riboswitches by using a tRNA as its effector.
stems of tRNA in the A and P sites move 50 to 30 , resulting in the
This riboswitch is believed to monitor the aminoacylation state of
hybrid A/P and P/E congurations [89]. This tRNA reorganisation is
tRNAs and to control the expression of genes involved in a number
102 ndez-Milla
P. Ferna n et al. / Archives of Biochemistry and Biophysics 602 (2016) 95e105

of functions related to aminoacylation and the metabolism of The precise nature of the factors involved in this process and their
amino acids [96]. The crystal structure of the highly conserved T- diversity throughout eukaryotes is only beginning to be unravelled.
box riboswitch in complex with tRNA has revealed how the distal In yeast, the tRNALys1 is shuttled to the mitochondrial surface by the
stem I region of the riboswitch interacts with the tRNA corner to glycolytic enzyme enolase 2 and taken over by the precursor of the
anchor it to the riboswitch [57]. mitochondrial lysyl-tRNA synthetase to be co-imported in mito-
Beyond their canonical fold, the three-dimensional architecture chondria through the protein import machinery [103]. In try-
of tRNAs is also found in a variety of other RNA molecules. These panosomes, tRNAs interact with the cytosolic elongation factor
structures, called tRNA-like (TLS), are often found in 50 and 30 eEF1a to reach the mitochondrial surface. Then, two mitochondrial
translated regions of plant mitochondrial mRNAs. They appear to membrane proteins, Tb11.01.4590 and Tb09.v1.0420, together with
serve as RNA processing signals for the maturation of mRNAs as Tim17 and Hsp70, are part of subsequent tRNA import machinery in
they are recognized and cleaved by RNase P and RNase Z enzymes the inner membrane showing that tRNA and protein import
[97,98]. Similarly, a number of positive strand RNA plant viral ge- mechanisms share components [104]. Finally, in plants some TOM
nomes also possess 3'-tRNA-like structures. Although differing proteins are believed to act as tRNA receptors on the mitochondrial
considerably from canonical tRNAs in terms of secondary structure, surface, while the voltage dependent anion channel (VDAC) may be
these TLS exhibit a three-dimensional fold that accurately mimics the main translocation channel through the outer mitochondrial
tRNAs and tRNA properties, including the capacity to be amino- membrane [105]. In all of these systems tRNAs interact with b-
acylated. These viral TLS often appear to be essential for viral barrel forming proteins such as VDAC, TOM40 or other components
replication [95]. For instance, the turnip yellow mosaic virus of the protein import machinery, presumably allowing them to
(TYMV) TLS (i) is recognized by ValRS like a true tRNA to amino- cross mitochondrial membranes. In all cases, structural modeling
acylate the viral genomic RNA and (ii) binds the promoter region predicts that mature tRNA have to undergo either complete or
in the 30 untranslated region of the TYMV RNA for negative-strand partial unfolding, or at least major structural rearrangements in
synthesis. The determination of the TYMV TLS three-dimensional order to cross mitochondrial membranes. The determination of
structure has revealed its global tRNA mimicry and has also structures of these factors in complex with tRNA will reveal the
shown how it exploits a completely different array of intra- mechanism of tRNA import at atomic resolution and, more gener-
molecular interactions to resemble tRNA. These specic in- ally, will help scientists to understand how tRNA import can be
teractions enable the TLS to switch conformations as required for ne-tuned to attain the pool of tRNA required for mitochondrial
its function [61]. translation.

6. Perspectives and new challenges of tRNA structural biology 6.3. Understanding the biogenesis and functions of tRNA fragments

6.1. Understanding the conformational dynamics of tRNAs Transfer RNAs are key players in a variety of molecular pro-
cesses, including, but not limited to, translation. For these crucial
The gallery of tRNA structures, either as isolated molecules or in functions, tRNAs have to be quality-controlled and their turn-over
complex with cellular partners (Fig. 5), highlights the exibility of regulated. Examples of this regulation include two tRNA surveil-
the L-shape scaffold. However, the cloverleaf behind this L-shape is lance pathways identied in yeast that degrade hypo-modied or
not always the only 2D conformation possible, nor the most stable mis-processed pre-tRNAs, as well as mature tRNAs lacking modi-
one. Defects in tRNA modication or mutations in mitochondrial cations. A nuclear tRNA surveillance pathway involving poly-
tRNAs related to human pathologies were for instance shown to adenylation by the TRAMP complex and degradation by the
destabilize the cloverleaf and favor alternate 2D folds for which 3D exosome deals primarily with hypo-modied tRNAs, while another
models still need to be established [26,99]. Also, atypical tRNAs pathway termed the rapid tRNA decay pathway involves a CCAse
found in animal mitochondria, with sequence alterations ranging and degradation by Xrn1 [106].
from a shortening of D/T loops, to the absence of D-arm, T-arm or Beyond tRNA degradation pathways, a number of other cellular
both arms, deserve further investigation to understand their activities generate tRNA-derived fragments (tRFs). The wide accu-
folding and their stability and to understand how translation fac- mulation of numerous tRFs suggests that these tRNA cleavage
tors can cope with such unusual and presumably exible sub- products potentially represent a new class of small non-coding
strates. The example of the human mitochondrial aspartyl-tRNA RNAs that may have important regulatory functions. tRFs are
synthetase suggests that the enzyme has relaxed its specicity and evolutionarily widespread and created by cleavage at a number of
shows a higher plasticity than its E. coli homolog, an evolutionary specic positions most probably by different endonuclease activ-
response to the structural alteration of its cognate tRNA [100]. ities (Fig. 2D). The discovery of small non-coding RNAs such as
microRNA (miRNA) and small interfering RNA (siRNA) and their
6.2. Understanding tRNA trafcking in the cell functions as major regulators of gene expression has been a major
breakthrough [107]. Recent analysis of transcriptomes in all three
In eukaryotes, gene expression is not restricted to the nucleus. It domains of life has suggested that tRFs constitute another class of
also takes place in organelles, in particular in mitochondria. The small regulatory RNAs with biologically relevant functions because
pool of tRNAs required for mitochondrial translation can be (i) in most cases tRFs are not generated from abundant tRNAs, and
composed of both tRNAs encoded in mitochondria and in the nu- the amount of tRFs does not correlate with initial tRNA abundance
cleus, with the latter being expressed in the cytosol and imported or with the corresponding tRNA gene copy number. (ii) The cleav-
to mitochondria by precise import systems. tRNA import is ubiq- age patterns of tRFs seem to be dependent on tRNA anticodons. (iii)
uitous in eukaryotes, although very diverse. In yeast and in some Cleavage patterns of tRNAs often vary according to the cell status
animals, a full set of tRNA is encoded in mitochondria and tRNA and/or developmental stages. (iv) Some tRFs have been found in
import is not required a priori but nevertheless takes place [101]. In association with key components of the RNA silencing machineries
contrast, in trypanosomes, not a single tRNA is encoded in the such as Argonaute or Piwi. The functions of tRFs are related to a
mitochondrial genome and the full set of tRNA is imported. An rapidly growing number of molecular processes, including trans-
intermediate situation is found in plants where a number of tRNAs lation, translation initiation, RNA degradation, reverse transcrip-
are not encoded in mitochondria and have to be imported [102]. tion, gene silencing, vacuolar ribophagy and cell survival [108]. The
P. Ferna n et al. / Archives of Biochemistry and Biophysics 602 (2016) 95e105 103

determination of three-dimensional structures of tRFs alone and in University of Strasbourg, an ANR Blanc research grant PRO-RNase
complex with protein partners will aid in understanding how tRFs P (ANR-11-BSV8-008-01/02) to PG and CS and by the LabEx con-
are generated as well as the mode of action and functional diversity sortium MitoCross in the frame of the French National Program
of this novel class of non coding RNAs [108,109]. Investissements dAvenir (ANR-11-LABX-0057_MITOCROSS).

6.4. From tRNA to systems biology References

The early observation of tRNAs capable of decoding nonsense [1] R. Giege, C. Sauter, Biocrystallography: past, present, future, HFSP J. 4 (2010)
UGA, UAA and UAG codons in E. coli [110], the reattribution of UGA [2] J.D. Watson, F.H.C. Crick, Molecular structure of nucleic acids: A structure for
to Phe in yeast mitochondria [111], and more recently the discovery deoxyribose nucleic acid, Nature 171 (1953) 737e738.
of the 21st and 22nd amino acids, selenocysteine (Sec) and pyro- [3] F.H.C. Crick, L. Barnett, S. Brenner, R.J. Watts-Tobin, General nature of the
genetic code for proteins, Nature 192 (1961) 1227e1232.
lysine (Pyl), coded by of UGA and UAG [112,113], have contributed to
[4] J.H. Matthaei, O.W. Jones, R.G. Martin, M.W. Nirenberg, Characteristics and
the idea of utilizing nonsense codons to incorporate non-canonical composition of RNA coding units, Proc. Natl. Acad. Sci. U. S. A. 48 (1962)
amino acids (ncAAs) into proteins. The expansion of genetic code 666e677.
in vivo requires the introduction of an orthogonal pair consisting of [5] F. Chapeville, F. Lipmann, G. von Ehrenstein, B. Weisblum, W.J. Ray, S. Benzer,
On the role of soluble ribonucleic acid in coding for amino acids, Proc. Natl.
a suppressor tRNA decoding a nonsense codon and a modied Acad. Sci. U. S. A. 48 (1962) 1086e1092.
cognate synthetase able to transfer the desired ncAA without [6] R.W. Holley, J. Apgar, G.A. Everett, J.T. Madison, M. Marquisee, S.H. Merrill,
interfering with tRNAs of the host organism. This has been achieved J.R. Penswick, A. Zamir, Structure of a ribonucleic acid, Science 147 (1965)
in a variety of ways, including the selection of variants of archeal [7] J.T. Madison, G.A. Everett, H. Kung, Nucleotide sequence of a yeast tyrosine
TyrRS [114,115] or PylRS [116,117] activating ncAAs. For optimal transfer RNA, Science 153 (1966) 531e534.
incorporation further engineering of host organism factors such as [8] U.L. RajBhandary, S.H. Chang, A. Stuart, R.D. Faulkner, R.M. Hoskinson,
H.G. Khorana, Studies on polynucleotides, LXVIII: the primary structure of
the elongation and release factors, or the peptidyl-transferase site yeast phenylalanine transfer RNA, Proc. Natl. Acad. Sci. U. S. A. 57 (1967) 751.
of the ribosome, may be necessary. More than one hundred such [9] H.G. Zachau, D. Dtting, H. Feldmann, The structures of two serine transfer
compounds have already been successfully incorporated into pro- ribonucleic acids. Hoppe-Seylers Z, Fr Physiol. Chem. 347 (1966) 212e235.
[10] S.H. Kim, F.L. Suddath, G.J. Quigley, A. McPherson, J.L. Sussman, A.H.J. Wang,
teins, introducing new functional groups for cross-linking or uo- N.C. Seeman, A. Rich, Three-dimensional tertiary structure of yeast phenyl-
rescent labeling purposes or to mimic epigenetic modications alanine transfer RNA, Science 185 (1974) 435e440.
[118,119]. In an alternative approach, RNA aptamers known as [11] J.D. Robertus, J.E. Ladner, J.T. Finch, D. Rhodes, R.S. Brown, B.F.C. Clark,
A. Klug, Structure of yeast phenylalanine tRNA at 3 resolution, Nature 250
exizymes have been designed to aminoacylate any tRNA with
(1974) 546e551.
ncAAs in cell free translation systems [120]. This system can be [12] C.D. Stout, H. Mizuno, J. Rubin, T. Brennan, S.T. Rao, M. Sundaralingam,
used to reprogram the code and introduce several ncAAs in a Atomic coordinates and molecular conformation of yeast phenylalanyl tRNA.
protein. These emerging strategies of code expansion or reprog- An independent investigation, Nucleic Acids Res. 3 (1976) 1111e1123.
[13] D. Moras, M.B. Comarmond, J. Fischer, R. Weiss, J.C. Thierry, J.P. Ebel, R. Giege
raming by reassigning rare codons to ncAAs open up vast per- Crystal structure of yeast tRNA(Asp), Nature 288 (1980) 669e674.
spectives for the production of articial proteins or more generally [14] M. Ibba, C. Francklyn, S. Cusack, Aminoacyl-tRNA Synthetases (Goergetown,
for the design of new organisms in the eld of synthetic biology TX: Landes Biosciences), 2005.
[15] A. Korostelev, S. Trakhanov, M. Laurberg, H.F. Noller, Crystal structure of a
[121,122]. 70S ribosome-tRNA complex reveals functional interactions and rearrange-
ments, Cell 126 (2006) 1065e1077.
7. Concluding remarks [16] M.Z. Barciszewska, P.M. Perrigue, J. Barciszewski, tRNA e the golden stan-
dard in molecular biology, Mol. Biosyst. 12 (2016) 12e17.
[17] B.P. Doctor, J. Apgar, R.W. Holley, Fractionation of yeast amino acid-acceptor
Forty years after the determination of the rst tRNA crystal ribonucleic acids by countercurrent distribution, J. Biol. Chem. 236 (1961)
structure, the adaptor molecule has not yet revealed all its secrets. 1117e1120.
[18] A. Hampel, M. Labanauskas, P.G. Connors, L. Kirkegard, U.L. Rajbhandary,
While at the end of the 90s, investigations carried out with model
P.B. Sigler, R.M. Bock, Single crystals of transfer RNA from formylmethionine
organisms may have given the feeling that the major aspects of and phenylalanine transfer RNA's, Science 162 (1968) 1384e1387.
tRNA:synthetase interactions had been revealed, the systematic [19] R. Giege, E. Touze, B. Lorber, A. Theobald-Dietrich, C. Sauter, Crystallogenesis
trends of free and liganded aminoacyl-tRNA synthetases, Cryst. Growth Des.
exploration of genomes since the early 2000 and an increasing
8 (2008) 4297e4306.
interest in pathology-related mutations, both in tRNAs and asso- [20] M. Levitt, Detailed molecular model for transfer ribonucleic acid, Nature 224
ciated proteins, have revitalized the eld and led to the discovery of (1969) 759e763.
many new aspects of tRNA biology. With the latest progress in [21] J. Witz, 1964: the rst model for the shape of a transfer RNA molecule. An
account of an unpublished small-angle X-ray scattering study, Biochimie 85
cryoEM, one can easily anticipate a burst of new data describing (2003) 1265e1268.
how tRNAs move on the ribosome and how tRNA dynamics con- [22] F. Jhling, M. Mo rl, R.K. Hartmann, M. Sprinzl, P.F. Stadler, J. Ptz, tRNAdb
tributes to the decoding of mRNA from bacteria to higher eukary- 2009: compilation of tRNA sequences and tRNA genes, Nucleic Acids Res. 37
(2009) D159eD162.
otes, including cytosolic, mitochondrial and plastidal translation [23] S. Wende, E.G. Platzer, F. Jhling, J. Ptz, C. Florentz, P.F. Stadler, M. Mo rl,
machineries. The integration of structural data from high resolution Biological evidence for the world's smallest tRNAs, Biochimie 100 (2014)
crystallography, cryoEM and correlative microscopy in the cell will 151e158.
[24] T. Suzuki, T. Suzuki, A complete landscape of post-transcriptional modica-
provide a much deeper insight into their distribution and functions tions in mammalian mitochondrial tRNAs, Nucleic Acids Res. 42 (2014)
with applications in human medicine, drug design and synthetic 7346e7357.
biology, making the eld of tRNA structural biology more active and [25] B. El Yacoubi, M. Bailly, V. de Cre cy-Lagard, Biosynthesis and function of
posttranscriptional modications of transfer RNAs, Annu. Rev. Genet. 46
open than ever. (2012) 69e95.
[26] M. Helm, R. Giege , C. Florentz, A WatsonCrick base-pair-disrupting methyl
Acknowledgments group (m1A9) is sufcient for cloverleaf folding of human mitochondrial
tRNA-Lys, Biochemistry 38 (1999) 13338e13346.
[27] W.A. Cantara, P.F. Crain, J. Rozenski, J.A. McCloskey, K.A. Harris, X. Zhang,
The authors wish to acknowledge the contribution of Dr Richard F.A.P. Vendeix, D. Fabris, P.F. Agris, The RNA modication database,
 in sharing his expertise on the history of tRNA biology and
Giege RNAMDB: 2011 update, Nucleic Acids Res. 39 (2011) D195eD201.

[28] M.A. Machnicka, K. Milanowska, O.O. Oglou, E. Purta, M. Kurkowska,
personal research artifacts, such as tRNA models, photographs, and
A. Olchowik, W. Januszewski, S. Kalinowski, S. Dunin-Horkawicz,
printed electron density maps. This work was supported by the K.M. Rother, et al., MODOMICS: a database of RNA modication
French Centre National de la Recherche Scientique, by the pathwaysd2013 update, Nucleic Acids Res. 41 (2013) D262eD267.
104 ndez-Milla
P. Ferna n et al. / Archives of Biochemistry and Biophysics 602 (2016) 95e105

[29] M.A. Rould, J.J. Perona, D. Soll, T.A. Steitz, Structure of E. coli glutaminyl-tRNA PLoS ONE 6 (2011) e20032.
synthetase complexed with tRNA-Gln and ATP at 2.8 resolution, Science [61] T.M. Colussi, D.A. Costantino, J.A. Hammond, G.M. Ruehle, J.C. Nix, J.S. Kieft,
246 (1989) 1135e1142. The structural basis of transfer RNA mimicry and conformational plasticity
[30] M. Ruff, S. Krishnaswamy, M. Boeglin, A. Poterszman, A. Mitschler, by a viral RNA, Nature 511 (2014) 366e369.
A. Podjarny, B. Rees, J.C. Thierry, D. Moras, Class II aminoacyl transfer RNA [62] M. Blaise, M. Bailly, M. Frechin, M.A. Behrens, F. Fischer, C.L. Oliveira,
synthetases: crystal structure of yeast aspartyl-tRNA synthetase complexed H.D. Becker, J.S. Pedersen, S. Thirup, D. Kern, Crystal structure of a transfer-
with tRNA (Asp), Science 252 (1991) 1682e1689. ribonucleoprotein particle that promotes asparagine formation, EMBO J. 29
[31] G. Eriani, M. Delarue, O. Poch, J. Gangloff, D. Moras, Partition of tRNA syn- (2010) 3118e3129.
thetases into two classes based on mutually exclusive sets of sequence [63] P. Nissen, M. Kjeldgaard, S. Thirup, G. Polekhina, L. Reshetnikova, B.F. Clark,
motifs, Nature 347 (1990) 203e206. J. Nyborg, Crystal structure of the ternary complex of Phe-tRNA(Phe), EF-Tu,
[32] R. Giege, M. Sissler, C. Florentz, Universal rules and idiosyncratic features in and a GTP analog, Science 270 (1995) 1464e1472.
tRNA identity, Nucleic Acids Res. 26 (1998) 5017e5035. [64] E. Schmitt, M. Panvert, C. Lazennec-Schurdevin, P.-D. Coureux, J. Perez,
[33] M.A. Rould, J.J. Perona, T.A. Steitz, Structural basis of anticodon loop recog- A. Thompson, Y. Mechulam, Structure of the ternary initiation complex
nition by glutaminyl-tRNA synthetase, Nature 352 (1991) 213e218. aIF2eGDPNPemethionylated initiator tRNA, Nat. Struct. Mol. Biol. 19 (2012)
[34] M. Naganuma, S. Sekine, Y.E. Chong, M. Guo, X.-L. Yang, H. Gamper, Y.- 450e454.
M. Hou, P. Schimmel, S. Yokoyama, The selective tRNA aminoacylation [65] T.M. Schmeing, R.M. Voorhees, A.C. Kelley, Y.-G. Gao, F.V. Murphy, J.R. Weir,
mechanism based on a single G$U pair, Nature 510 (2014) 507e511. V. Ramakrishnan, The crystal structure of the ribosome bound to EF-Tu and
[35] V. Biou, A. Yaremchuk, M. Tukalo, S. Cusack, The 2.9 crystal structure of aminoacyl-tRNA, Science 326 (2009) 688e694.
T. thermophilus seryl-tRNA synthetase complexed with tRNA(Ser), Science [66] O. Pellegrini, I. Li de la Sierra-Gallay, J. Piton, L. Gilet, C. Condon, Activation of
263 (1994) 1404e1410. tRNA maturation by downstream uracil residues in B. subtilis, Structure 20
[36] S. Hauenstein, C.-M. Zhang, Y.-M. Hou, J.J. Perona, Shape-selective RNA (2012) 1769e1777.
recognition by cysteinyl-tRNA synthetase, Nat. Struct. Mol. Biol. 11 (2004) [67] R. Ishitani, O. Nureki, N. Nameki, N. Okada, S. Nishimura, S. Yokoyama,
1134e1141. Alternative tertiary structure of tRNA for recognition by a posttranscriptional
[37] R. Giege, Toward a more complete view of tRNA biology, Nat. Struct. Mol. modication enzyme, Cell 113 (2003) 383e394.
Biol. 15 (2008) 1007e1014. [68] K. Tomita, S. Fukai, R. Ishitani, T. Ueda, N. Takeuchi, D.G. Vassylyev, O. Nureki,
[38] M. Guo, P. Schimmel, Structural analyses clarify the complex control of Structural basis for template-independent RNA polymerization, Nature 430
mistranslation by tRNA synthetases, Curr. Opin. Struct. Biol. 22 (2012) (2004) 700e704.
119e126. [69] F. Pinker, G. Bonnard, A. Gobert, B. Gutmann, K. Hammani, C. Sauter,
[39] J.A. Abbott, C.S. Francklyn, S.M. Robey-Bond, Transfer RNA and human dis- P.A. Gegenheimer, P. Giege , PPR proteins shed a new light on RNase P
ease, Front. Genet. 5 (2014) 158. biology, RNA Biol. 10 (2013) 1457e1468.
[40] H. Schwenzer, J. Zoll, C. Florentz, M. Sissler, Pathogenic implications of hu- [70] A. Gobert, F. Pinker, O. Fuchsbauer, B. Gutmann, R. Boutin, P. Roblin,
man mitochondrial aminoacyl-tRNA synthetases, Top. Curr. Chem. 344 C. Sauter, P. Giege , Structural insights into protein-only RNase P complexed
(2014) 247e292. with tRNA, Nat. Commun. 4 (2013) 1353.
[41] F. Michel, E. Westhof, Modelling of the three-dimensional architecture of [71] H. Betat, Y. Long, J.E. Jackman, M. Mo rl, From end to end: tRNA editing at 5-
group I catalytic introns based on comparative sequence analysis, J. Mol. Biol. and 3-terminal positions, Int. J. Mol. Sci. 15 (2014) 23975e23998.
216 (1990) 585e610. [72] A. Guelorget, B. Golinelli-Pimpaneau, Mechanism-based strategies for trap-
[42] J.H. Cate, A.R. Gooding, E. Podell, K. Zhou, B.L. Golden, C.E. Kundrot, et al., ping and crystallizing complexes of RNA-modifying enzymes, Structure 19
Crystal structure of a group I ribozyme domain: principles of RNA packing, (2011) 282e291.
Science 273 (1996) 1678e1684. [73] S. Goto-Ito, T. Ito, M. Kuratani, Y. Bessho, S. Yokoyama, Tertiary structure
[43] H.W. Pley, D.S. Lindes, C. DeLuca-Flaherty, D.B. McKay, Crystals of checkpoint at anticodon loop modication in tRNA functional maturation,
hammerhead ribozyme, J. Biol. Chem. 268 (1993) 19656e19658. Nat. Struct. Mol. Biol. 16 (2009) 1109e1115.
[44] W.G. Scott, J.T. Finch, R. Grenfell, J. Fogg, T. Smith, M.J. Gait, A. Klug, Rapid [74] B. Pan, Y. Xiong, T.A. Steitz, How the CCA-adding enzyme selects adenine
crystallization of chemically synthesized Hammerhead RNAs using a double over cytosine at position 76 of tRNA, Science 330 (2010) 937e940.
screening procedure, J. Mol. Biol. 250 (1995) 327e332. [75] V. Ramakrishnan, The ribosome emerges from a black box, Cell 159 (2014)
[45] N. Ban, P. Nissen, J. Hansen, P.B. Moore, T.A. Steitz, The complete atomic 979e984.
structure of the large ribosomal subunit at 2.4 resolution, Science 289 [76] X. Agirrezabala, M. Valle, Structural insights into tRNA dynamics on the
(2000) 905e920. ribosome, Int. J. Mol. Sci. 16 (2015) 9866e9895.
[46] B.T. Wimberly, D.E. Brodersen, W.M. Clemons, R.J. Morgan-Warren, [77] S. Uemura, C.E. Aitken, J. Korlach, B.A. Flusberg, S.W. Turner, J.D. Puglisi, Real-
A.P. Carter, C. Vonrhein, T. Hartsch, V. Ramakrishnan, Structure of the 30S time tRNA transit on single translating ribosomes at codon resolution, Na-
ribosomal subunit, Nature 407 (2000) 327e339. ture 464 (2010) 1012e1017.
[47] M.M. Yusupov, G.Z. Yusupova, A. Baucom, K. Lieberman, T.N. Earnest, [78] E. Schmitt, M. Panvert, S. Blanquet, Y. Mechulam, Crystal structure of
J.H. Cate, H.F. Noller, Crystal structure of the ribosome at 5.5 resolution, methionyl-tRNAfMet transformylase complexed with the initiator formyl-
Science 292 (2001) 883e896. methionyl-tRNAfMet, EMBO J. 17 (1998) 6819e6826.
[48] D.J. Klein, T.M. Schmeing, P.B. Moore, T.A. Steitz, The kink-turn: a new RNA [79] P. Barraud, E. Schmitt, Y. Mechulam, F. Dardel, C. Tisne, A unique confor-
secondary structure motif, EMBO J. 20 (2001) 4214e4221. mation of the anticodon stem-loop is associated with the capacity of
[49] M. Meyer, E. Westhof, B. Masquida, A structural module in RNase P expands tRNAfMet to initiate protein synthesis, Nucleic Acids Res. 36 (2008)
the variety of RNA kinks, RNA Biol. 9 (2012) 254e260. 4894e4901.
[50] A. Lescoute, N.B. Leontis, C. Massire, E. Westhof, Recurrent structural RNA [80] A. Tsai, A. Petrov, R.A. Marshall, J. Korlach, S. Uemura, J.D. Puglisi, Hetero-
motifs, isostericity matrices and sequence alignments, Nucleic Acids Res. 33 geneous pathways and timing of factor departure during translation initia-
(2005) 2395e2409. tion, Nature 487 (2012) 390e393.
[51] L. Jaeger, E.J. Verzemnieks, C. Geary, The UA_handle: a versatile submotif in [81] N. Sonenberg, A.G. Hinnebusch, Regulation of translation initiation in eu-
stable RNA architectures, Nucleic Acids Res. 37 (2009) 215e230. karyotes: mechanisms and biological targets, Cell 136 (2009) 731e745.
[52] E.A. Doherty, R.T. Batey, B. Masquida, J.A. Doudna, A universal mode of helix [82] A.G. Hinnebusch, The scanning mechanism of eukaryotic translation initia-
packing in RNA, Nat. Struct. Biol. 8 (2001) 339e343. tion, Annu. Rev. Biochem. 83 (2014) 779e812.
[53] P. Nissen, J.A. Ippolito, N. Ban, P.B. Moore, T.A. Steitz, RNA tertiary in- n, P. Milon, X. Agirrezabala, G. Lasso, D. Gil, M.V. Rodnina, M. Valle, The
[83] P. Julia
teractions in the large ribosomal subunit: the A-minor motif, Proc. Natl. cryo-EM structure of a complete 30S translation initiation complex from
Acad. Sci. U. S. A. 98 (2001) 4899e4903. Escherichia coli, PLoS Biol. 9 (2011) e1001095.
[54] B. Masquida, B. Beckert, F. Jossinet, Exploring RNA structure by integrative [84] J.M. Ogle, F.V. Murphy IV, M.J. Tarry, V. Ramakrishnan, Selection of tRNA by
molecular modelling, New Biotechnol. 27 (2010) 170e183. the ribosome requires a transition from an open to a closed form, Cell 111
[55] M. Meyer, H. Nielsen, V. Olie ric, P. Roblin, S.D. Johansen, E. Westhof, (2002) 721e732.
B. Masquida, Speciation of a group I intron into a lariat capping ribozyme, [85] N. Demeshkina, L. Jenner, E. Westhof, M. Yusupov, G. Yusupova, A new un-
Proc. Natl. Acad. Sci. U. S. A. 111 (2014) 7659e7664. derstanding of the decoding principle on the ribosome, Nature 484 (2012)
[56] N.J. Reiter, A. Osterman, A. Torres-Larios, K.K. Swinger, T. Pan, A. Mondrago  n, 256e259.
Structure of a bacterial ribonuclease P holoenzyme in complex with tRNA, [86] T.M. Schmeing, V. Ramakrishnan, What recent ribosome structures have
Nature 468 (2010) 784e789. revealed about the mechanism of translation, Nature 461 (2009) 1234e1242.
[57] J. Zhang, A.R. Ferre-DAmare , Co-crystal structure of a T-box riboswitch stem [87] C.E. Aitken, J.D. Puglisi, Following the intersubunit conformation of the
I domain in complex with its cognate tRNA, Nature 500 (2013) 363e366. ribosome during translation in real time, Nat. Struct. Mol. Biol. 17 (2010)
[58] J. Lehmann, F. Jossinet, D. Gautheret, A universal RNA structural motif 793e800.
docking the elbow of tRNA in the ribosome, RNAse P and T-box leaders, [88] S. Kuhlenkoetter, W. Wintermeyer, M.V. Rodnina, Different substrate-
Nucleic Acids Res. 41 (2013) 5494e5502. dependent transition states in the active site of the ribosome, Nature 476
[59] C.Y. Cheng, F.-C. Chou, W. Kladwang, S. Tian, P. Cordero, R. Das, Consistent (2011) 351e354.
global structures of complex RNA states through multidimensional chemical [89] D. Moazed, H.F. Noller, Intermediate states in the movement of transfer RNA
mapping, eLife 4 (2015) e07600. in the ribosome, Nature 342 (1989) 142e148.
[60] O.M. Ganichkin, E.A. Anedchenko, M.C. Wahl, Crystal structure analysis re- [90] X. Agirrezabala, H.Y. Liao, E. Schreiner, J. Fu, R.F. Ortiz-Meoz, K. Schulten,
veals functional exibility in the selenocysteine-specic tRNA from mouse, R. Green, J. Frank, Structural characterization of mRNA-tRNA translocation
P. Ferna n et al. / Archives of Biochemistry and Biophysics 602 (2016) 95e105 105

intermediates, Proc. Natl. Acad. Sci. 109 (2012) 6094e6099. [108] C. Megel, G. Morelle, S. Lalande, A.-M. Duche ^ne, I. Small, L. Marechal-
[91] N. Fischer, A.L. Konevega, W. Wintermeyer, M.V. Rodnina, H. Stark, Ribosome Drouard, Surveillance and cleavage of eukaryotic tRNAs, Int. J. Mol. Sci. 16
dynamics and tRNA movement by time-resolved electron cryomicroscopy, (2015) 1873e1893.
Nature 466 (2010) 329e333. [109] A. Kanai, Disrupted tRNA genes and tRNA fragments: a perspective on tRNA
[92] A. Amunts, A. Brown, J. Toots, S.H. Scheres, V. Ramakrishnan, The structure of gene evolution, Life 5 (2015) 321e331.
the human mitochondrial ribosome, Science 348 (2015) 95e98. [110] G. Eggertsson, D. So ll, Transfer ribonucleic acid-mediated suppression of
[93] B.J. Greber, P. Bieri, M. Leibundgut, A. Leitner, R. Aebersold, D. Boehringer, termination codons in Escherichia coli, Microbiol. Rev. 52 (1988) 354e374.
N. Ban, The complete structure of the 55S mammalian mitochondrial ribo- [111] G. Macino, G. Coruzzi, F.G. Nobrega, M. Li, A. Tzagoloff, Use of the UGA
some, Science 348 (2015) 303e308. terminator as a tryptophan codon in yeast mitochondria, Proc. Natl. Acad.
[94] S.F.J. Le Grice, In the beginning: initiation of minus strand DNA synthesis in Sci. U. S. A. 76 (1979) 3784e3785.
retroviruses and LTR-containing retrotransposons, Biochemistry 42 (2003) [112] A. Scho n, A. Bo
ck, G. Ott, M. Sprinzl, D. Soll, The selenocysteine-inserting opal
14349e14355. suppressor serine tRNA from E. coli is highly unusual in structure and
[95] T.W. Dreher, Viral tRNAs and tRNA-like structures, Wiley Interdiscip. Rev. modication, Nucleic Acids Res. 17 (1989) 7159e7165.
RNA 1 (2010) 402e414. [113] G. Srinivasan, C.M. James, J.A. Krzycki, Pyrrolysine encoded by UAG in
[96] J.C. Grigg, A. Ke, Sequence, structure, and stacking: specics of tRNA Archaea: charging of a UAG-decoding specialized tRNA, Science 296 (2002)
anchoring to the T box riboswitch, RNA Biol. 10 (2013) 1761e1764. 1459e1462.
[97] A. Gobert, B. Gutmann, A. Taschner, M. Go ringer, J. Holzmann, [114] K. Sakamoto, A. Hayashi, A. Sakamoto, D. Kiga, H. Nakayama, A. Soma,
R.K. Hartmann, W. Rossmanith, P. Giege , A single Arabidopsis organellar T. Kobayashi, M. Kitabatake, K. Takio, K. Saito, et al., Site-specic incorpo-
protein has RNase P activity, Nat. Struct. Mol. Biol. 17 (2010) 740e744. ration of an unnatural amino acid into proteins in mammalian cells, Nucleic
[98] B. Gutmann, A. Gobert, P. Giege , PRORP proteins support RNase P activity in Acids Res. 30 (2002) 4692e4699.
both organelles and the nucleus in Arabidopsis, Genes Dev. 26 (2012) [115] L. Wang, T.J. Magliery, D.R. Liu, P.G. Schultz, A new functional suppressor
1022e1027. tRNA/aminoacyl-tRNA synthetase pair for the in vivo incorporation of un-
[99] M. Sissler, M. Helm, M. Frugier, R. Giege , C. Florentz, Aminoacylation prop- natural amino acids into proteins, J. Am. Chem. Soc. 122 (2000) 5010e5011.
erties of pathology-related human mitochondrial tRNA-Lys variants, RNA 10 [116] C.R. Polycarpo, S. Herring, A. Be rube , J.L. Wood, D. Soll, A. Ambrogelly, Pyr-
(2004) 841e853. rolysine analogues as substrates for pyrrolysyl-tRNA synthetase, FEBS Lett.
[100] A. Neuenfeldt, B. Lorber, E. Ennifar, A. Gaudry, C. Sauter, M. Sissler, 580 (2006) 6695e6700.
C. Florentz, Thermodynamic properties distinguish human mitochondrial [117] T. Yanagisawa, R. Ishii, R. Fukunaga, T. Kobayashi, K. Sakamoto, S. Yokoyama,
aspartyl-tRNA synthetase from bacterial homolog with same 3D architec- Multistep engineering of pyrrolysyl-tRNA synthetase to genetically encode
ture, Nucleic Acids Res. 41 (2013) 2698e2708. N3-(o-azidobenzyloxycarbonyl) lysine for site-specic protein modication,
[101] N.S. Entelis, S. Kieffer, O.A. Kolesnikova, R.P. Martin, I.A. Tarassov, Structural Chem. Biol. 15 (2008) 1187e1197.
requirements of tRNA-Lys for its import into yeast mitochondria, Proc. Natl. [118] J.A. Johnson, Y.Y. Lu, J.A. Van Deventer, D.A. Tirrell, Residue-specic incor-
Acad. Sci. 95 (1998) 2838e2843. poration of non-canonical amino acids into proteins: recent developments
[102] T. Salinas-Giege , R. Giege
, P. Giege, tRNA biology in mitochondria, Int. J. Mol. and applications, Curr. Opin. Chem. Biol. 14 (2010) 774e780.
Sci. 16 (2015) 4518e4559. [119] T. Mukai, T. Yanagisawa, K. Ohtake, M. Wakamori, J. Adachi, N. Hino, A. Sato,
[103] M. Baleva, A. Gowher, P. Kamenski, I. Tarassov, N. Entelis, B. Masquida, T. Kobayashi, A. Hayashi, M. Shirouzu, et al., Genetic-code evolution for
A moonlighting human protein is involved in mitochondrial import of tRNA, protein synthesis with non-natural amino acids, Biochem. Biophys. Res.
Int. J. Mol. Sci. 16 (2015) 9354e9367. Commun. 411 (2011) 757e761.
[104] D. Seidman, D. Johnson, V. Gerbasi, D. Golden, R. Orlando, S. Hajduk, Mito- [120] H. Murakami, A. Ohta, H. Ashigai, H. Suga, A highly exible tRNA acylation
chondrial membrane complex that contains proteins necessary for tRNA method for non-natural polypeptide synthesis, Nat. Methods 3 (2006)
import in Trypanosoma brucei, J. Biol. Chem. 287 (2012) 8892e8903. 357e359.
[105] T. Salinas, A.-M. Duche ^ne, L. Mare chal-Drouard, Recent advances in tRNA [121] D. Soll, A tRNA-guided research journey from synthetic chemistry to syn-
mitochondrial import, Trends Biochem. Sci. 33 (2008) 320e329. thetic biology, RNA 21 (2015) 742e744.
[106] J.E. Wilusz, J.M. Whipple, E.M. Phizicky, P.A. Sharp, tRNAs marked with [122] N. Terasaka, Y. Iwane, A.-S. Geiermann, Y. Goto, H. Suga, Recent de-
CCACCA are targeted for degradation, Science 334 (2011) 817e821. velopments of engineered translational machineries for the incorporation of
[107] S. Jonas, E. Izaurralde, Towards a molecular understanding of microRNA- non-canonical amino acids into polypeptides, Int. J. Mol. Sci. 16 (2015)
mediated gene silencing, Nat. Rev. Genet. 16 (2015) 421e433. 6513e6531.